Ebola optimization based spiking neural network for automatic hate speech recognition

Ebola optimization based spiking neural network for automatic hate speech recognition

연구 분야: Artificial Intelligence

학회: International Journal of Information Technology

In this paper, efficient machine learning technique is introduced to develop efficient machine learning model for hate speech recognition from the tweet data. Initially, the tweet data is gathered from the open-source system. Hence, to achieve hate speech recognition, different key tasks are considered which is named as pre-processing stage, feature extraction stage and classification stage. In the pre-processing stage, different pre-processing methods are considered for removing unwanted information from the tweet data such as Character Removal, unwanted symbols removal, URL and tag removal, irrelevant content removal, Stemming, tokenization, parts of speech (POS) and negation vector. The extracted features are utilized to identify the hate speech recognition from the tweet data. The proposed classifier is named as Spiking Neural Network- Ebola Optimization Algorithm (SNN-EOA). The proposed classifier is a combination of Spiking Neural Network (SNN) and Ebola Optimization Algorithm (EOA). The performance of proposed approach is analysed based on different metrics.